MIT Research Positions in Machine Learning for Health

Contact: Dr. Li-wei Lehman

I am looking for highly-motivated Research Fellows and Research Interns to join my team at the Institute for Medical Engineering & Science at MIT to work on projects in machine learning for health. The positions offer opportunities to develop machine learning and statistical methods to derive insights from multimodal, longitudinal observational health data for informed treatment decision making. This would be an ideal position for a candidate who has a strong machine learning (ML) background, and would like to conduct research in the intersection of ML for health, probabilistic generative modeling, causal inference, dynamic treatment regimes, and off-line model-based reinforcement learning. This position provides the opportunity to work with an interdisciplinary team of researchers with expertise in machine learning, causal inference, dynamic treatment regimes, and clinical medicine. Relevant Project Page: Learning Optimal Dynamic Treatment Regimes from Health Data.

Example topics of interests include (but not limited to):

Safe and Robust Machine Learning for Informed Sequential Treatment Decision Making
- Well-calibrated uncertainty estimates
- Constrained inference using domain knowledge
- Counterfactual prediction and explanations under dynamic and time-varying treatment regimes
Probabilistic Modeling and Variational Inference in Time Series Modeling for Health. See our AAAI 2022 paper on Knowledge Distillation via Constrained Variational Inference.
Sequential Treatment Outcome Prediction under Distribution Shift
Interpretable Offline Reinforcement Learning and Deep Probabilistic Time Series Models for Health and Sequential Treatment Decision Making
Representation Learning on Multimodal Data

Knowledge and experience in machine learning required. Knowledge in one or more of the following areas would be a plus: probabilistic models, generative models, deep learning, transfer and continual learning, causal machine learning, policy selection, and off-line model-based reinforcement learning. Duties will include conducting original research, publishing in top-tier machine learning conferences and scientific journals, mentoring students, and collaborating on research grant proposal writing.

Immediate openings are available for the following positions: Research Scientists/Research Intern/Research Assistant position -- this is a 3-6 months (December 2022 - May 2023) funded position for candidates who recently obtained a Ph.D. or are currently enrolled in an advanced degree program related to machine learning or statistical inference with a publication record in top AI or ML for health conferences. The ideal candidate will have demonstrated an outstanding capability for independent research and a solid publication record in top-tier machine learning, AI conferences or other top-tier ML for healthcare conferences and journals. Candidate must hold an advanced degree in Computer Science, Machine Learning, Statistics, or a related field.

Postdoc Fellow is for candidates who have secured independent funding covering all expenses (e.g., through external Fellowship programs). Ph.D. candidates who are enrolled in a top Computer Science program and have a strong publication record in AI conferences may apply for Research Intern positions. MIT undergraduate students or exchange students who are eligible to participate in the MIT UROP program: please email Dr. Lehman your resume and the semesters you are able to work as a UROP.

In addition to a curriculum vitae, applicants should submit links to 1-3 most relevant publications to Li-wei Lehman, lilehman<at>mit.edu ASAP. Please specify the possible date range to start the position.

Li-wei Lehman, Ph.D. 
Research Scientist 
Institute for Medical Engineering & Science
Massachusetts Institute of Technology
http://web.mit.edu/lilehman/www/